Cache unpacked Gradle wrapper distributions in GitHub Actions workflows

[ggrell]

Motivation

• CI runs were repeatedly downloading and decompressing the Gradle wrapper ZIPs because the unpacked wrapper distribution folder is not reliably preserved by actions/setup-java's cache: 'gradle' alone.
• Persisting the unpacked Gradle binaries across runs reduces download/decompress overhead and speeds builds for PR, main (snapshot), and release workflows.

Description

• Added an actions/cache@v4 step to pull_request, merge_master, and publish_release workflows to cache ~/.gradle/wrapper/dists so the unpacked Gradle distributions are restored between runs.
• The cache is keyed by ${{ runner.os }}-gradle-wrapper-dists-${{ hashFiles('**/gradle/wrapper/gradle-wrapper.properties') }} with an OS-level restore key fallback.
• Left actions/setup-java with cache: 'gradle' enabled so dependency/module caching and the new wrapper-distribution cache work together.
• Modified files: .github/workflows/pull_request.yml, .github/workflows/merge_master.yml, and .github/workflows/publish_release.yml.

Testing

• Parsed each updated workflow YAML with ruby -e "require 'yaml'; %w[.github/workflows/pull_request.yml .github/workflows/merge_master.yml .github/workflows/publish_release.yml].each{|f| YAML.load_file(f); puts \"OK #{f}\" }" and the parser reported success for all three files.
• Verified the workflow diffs with git diff to confirm the intended cache steps were added successfully.

---

Codex Task

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.11%. Comparing base (cfc13bc) to head (a58a535).
⚠️ Report is 10 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff            @@
##               main     #308   +/-   ##
=========================================
  Coverage     88.11%   88.11%           
  Complexity       48       48           
=========================================
  Files             9        9           
  Lines           101      101           
=========================================
  Hits             89       89           
  Misses           12       12           

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0471d5d...a58a535. Read the comment docs.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.